jarvisbox

Remove Image Background Online

Remove solid or near-solid colour backgrounds from images and export as transparent PNG. Works best on logos, illustrations, and product shots on clean studio backgrounds. The tool samples the background colour from the image corners and uses a flood-fill algorithm with adjustable tolerance to remove connected matching pixels.

100% in your browser. Your file never leaves your device.
Works best on images with a clean, solid-colour background (logos, illustrations, product shots on white). For complex photographic backgrounds use a dedicated AI tool.
Background removal settings

Load an image to begin.

How to use

  1. 1. Drop your image onto the upload zone or tap to select from your device.
  2. 2. Leave "Auto-detect" checked to sample background from corners, or uncheck to pick a colour manually.
  3. 3. Adjust tolerance — lower (10–20) for clean backgrounds, higher (40–60) for backgrounds with slight gradients.
  4. 4. Click "Remove background" to process.
  5. 5. Download the transparent PNG.

Frequently asked questions

What kind of images work best?

Images with a clean, solid-colour background: logos on white, product shots on grey or white studio backdrop, illustrations with flat-colour backgrounds. Complex photographic backgrounds are not well-suited to this approach.

What does the tolerance slider do?

Tolerance controls how similar a pixel must be to the background colour to be removed. Higher tolerance removes more pixels (useful for slightly noisy backgrounds), lower tolerance is more precise (avoids removing pixels near the subject).

Why is auto-detect sometimes wrong?

Auto-detect samples the four corner pixels and averages them. If your image has a border, logo corner elements, or subject that extends to the corners, the sampled "background" may be incorrect. Uncheck auto-detect and pick the background colour manually.

Can I remove a coloured background?

Yes. Uncheck auto-detect and use the colour picker to select the exact background colour. Set a moderate tolerance (20–40) for solid colours, higher for backgrounds with slight variation.

Why are some background areas not removed?

The flood-fill algorithm starts from the image corners and only removes pixels connected to the background. Enclosed background areas (surrounded by foreground) are not reached. This is by design — it prevents accidentally removing interior areas that match the background colour.

For complex photos, what should I use?

This tool uses a colour threshold approach suited for solid backgrounds. For complex photographic backgrounds (outdoor photos, varied textures), AI-powered tools (like remove.bg or Canva's background remover) use segmentation models and handle these cases much better.

反馈这个工具的问题